The same simulation urge runs in many game developers. Fight this urge. Some artists want their models to be perfect, adding finger joints even when the character will never be close enough to the camera to see his hands, much less fingers.

Some programmers like to model physics when simple rules are more fun. And designers may find detailed character backstories interesting, but that doesn't mean the backstory needs to be explained in a massive cutscene.

In each of these cases, the urge for simulation is at odds with the player's needs. And with game development, anything that is not directly serving the player is just wasted development effort.

Detail is only useful when it aids the game. As with D&D, most enemies are only on the screen for a few seconds. They only need enough actions to support that brief time, and nothing else.

Instead of adding detailed situational AI or extra animations that the player won't notice, focus on the one or two unique things that make that enemy stand out. And make those few things shine for the time the enemy is actually on screen.

What concepts does this enemy teach the player? What will the player remember later? Design around that. If the enemy appears over and over again throughout the game, make sure they do something different every now and then for variety, but don't add details just for detail's sake. Design for fun rather than simulation.

Flexible Systems

One of the bigger mechanical changes in 4E is Saving Throws. Rather than treat it as a unique mechanic onto itself as previous editions did, it's been combined into the standard defense mechanism. So Reflex, Fortitude and Will are just different flavors of Armor Class, providing four different defense stats that all work the same way.

Whenever a character does anything attack-like, whether swinging a sword, breathing fire, or subverting someone's mind, the player rolls an attack against one of these defense stats. And these numbers scale up like everything else.

This is a good example of a standardized system that allows easy adaptability. If the DM wants to invent a new monster attack, or a new trap, or even improvise a new action in the middle of a game, it's fairly easy to determine how to resolve it. Generally, roll an attack or ability and compare it to one of these defenses.

Having four defense stats means it's easy to generate attacks that feel unique and different but operate on the same basis mechanically. And by exposing the underlying math, a good DM knows how to generate unique content for each encounter and to adjudicate when something strange comes up, which it usually does.

This doesn't all translate directly to video games, since we lack a DM at the table to interpret these rules real-time, but it does give some good guidelines for feature and content design. Create systems with enough flexibility to cover a wide range of real situations.

Make sure the mechanics emphasize the differences between situations rather than flattening them all out to be the same -- in this example, every player has different values for each of the four defenses, so an easy way to vary encounters is to focus on different defenses.